BUSINESS
What to Consider Before Investing in Banking Shares: A Practical Checklist
Before investing in banking shares, match the dividend to spare capital, late loans, and price to book, because a low price can already reflect losses.
You have a bank share on a watchlist because the dividend yield looks generous, or because the price sits under book value. Banking shares are the equity in a lender. The bank takes in deposits, pays interest on some of them, and lends or invests the money at a higher rate, and your claim sits behind depositors and other creditors.
That gap between the rate earned and the rate paid is the business, and shareholder equity funds only a slice of the assets. Loan losses and markdowns on bonds land on owners before they land on depositors. A low price or a fat yield does not tell you whether that slice is thick enough to keep paying you.
Before investing in banking shares, read five things in the latest report: capital against the payout rule, late loans and the reserve against them, the interest margin, how the bank funds itself, and the price you are paying for tangible book. The checks below follow that order, which is the order that decides whether the dividend and the discount are still intact.
What You Own in a Bank Share
A normal company sells a product and keeps the profit after costs. A bank’s product is a loan, and the bank only wins if the borrower pays it back over years. Interest income is the main engine for a plain lender, while fees, cards, trading, and wealth management matter more at banks that are not mostly a loan book.
The funding is other people’s money. Deposits that pay little or no interest, such as many checking accounts, are the cheap source. When those leave and the bank replaces them with wholesale funds or high-rate savings, the margin shrinks, especially if the loans cannot reprice as fast as the deposits. Rapid loan growth funded that way is a different business from slow growth funded by sticky local accounts.
Your shares are not a claim on the deposits, and they are not a safer version of a savings account at the same bank. If losses use up equity, owners are diluted or wiped out while insured depositors still stand in a different line. That is why a bank can look profitable on a single quarter and still be a poor share to hold.
The Six-Part Rating Shareholders Cannot See
US bank examiners have rated institutions under the Uniform Financial Institutions Rating System since 1979, and the version in use scores six parts known as CAMELS. Each part is scored from 1 to 5, with 1 the strongest and 5 the weakest, and examiners also assign an overall rating from those parts. You will not be handed that score. National banks and federal savings associations are barred from disclosing it, under the rules barring disclosure of a CAMELS rating, and an unauthorized disclosure can bring criminal penalties.
This prohibition applies to any portion of a report of examination, supervisory correspondence, and any representations concerning the report or supervisory correspondence, or their findings, including the assigned CAMELS rating.
Office of the Comptroller of the Currency, Bulletin 2019-15, March 25, 2019
The public substitute is to rebuild the six questions from reports anyone can open. The Kansas City Fed’s plain account of how Federal Reserve examiners score banks is the map. You are not guessing the secret number. You are asking the same six questions with public figures.
- Capital: Equity that can absorb losses and still leave the bank above the ratios its regulator requires.
- Asset quality: Credit risk in loans, securities, foreclosed property, and commitments that sit off the balance sheet until a customer draws them.
- Management: Whether the board can spot those risks and keep the bank inside the law, which has no single ratio.
- Earnings: Whether profit is large enough, and whether it comes from repeatable spread and fees rather than a one-time gain.
- Liquidity: Cash and stable funding set against deposits and other payments the bank cannot skip.
- Market risk: How a move in interest rates, currencies, commodities, or equities would hit earnings or capital.
For a US bank, the public file is the quarterly call report, the FFIEC Uniform Bank Performance Report that lines those figures up, and any annual or quarterly filing with the SEC. Public enforcement orders are public too. A slide that quotes only the dividend is not that file. Banks outside the United States publish the same ideas under local names, in a capital report and a credit-quality note.
Capital Ratios That Gate the Dividend
Capital is the first check because it is the piece that can turn a dividend off even while the bank still looks profitable. Common equity tier 1, or CET1, is the core of that capital: mostly ordinary shares and retained earnings, divided by assets that have been weighted for risk. The minimum capital ratios in 12 CFR 324.10 set the US floor for an FDIC-supervised bank, and a separate conservation buffer, made of CET1, limits dividends, buybacks, and discretionary bonuses when the cushion over those floors is not full.
A second test, prompt corrective action, uses the phrase well capitalized for a higher set of floors. The two tests are not the same gate. A bank can clear the well-capitalized CET1 line and still be inside the conservation buffer, which is the detail people skip when they call a bank safe because it is well capitalized. The well capitalized tests in 12 CFR 324.403 are the source for that second column. Largest banks face extra charges on top, and some smaller banks elect a community bank leverage ratio instead of this whole stack.
US CAPITAL LINES FOR AN FDIC-SUPERVISED BANK
| Ratio | Minimum | Full base buffer | Well capitalized |
|---|---|---|---|
| CET1 | 4.5% | Above 7% | At least 6.5% |
| Tier 1 risk-based | 6% | Above 8.5% | At least 8% |
| Total risk-based | 8% | Above 10.5% | At least 10% |
| Leverage | 4% | Separate leverage buffer, if one applies | At least 5% |
The full-buffer column is the minimum plus 2.5 percentage points. The binding buffer is the smallest of those three gaps, and any countercyclical buffer the regulator has switched on sits on top. A bank whose cushion is not above that line must keep distributions inside a maximum payout for the quarter. Read the bank’s own required CET1 if it publishes one, because a global bank’s private target is often well above 7 percent, and the legal line in another country is not this table.
TWO SAMPLE CET1 RATIOS
| Sample CET1 | Gap over the 4.5% minimum | Full 2.5 point base buffer | Clears the 6.5% well-capitalized CET1 test |
|---|---|---|---|
| 6.8% | 2.3 points | No | Yes |
| 8.0% | 3.5 points | Yes | Yes |
These two rows are an illustration, not two real banks. The first row is the case buyers miss: the bank can be called well capitalized on the CET1 test and still face a cap on the dividend and on buybacks. The second row clears both the payout cushion and that well-capitalized test. If your bank reports 6.8 percent and talks up a rising dividend, the table is the objection.
How Problem Loans Show Up in the Accounts
Asset quality is the next check, because capital that looks fine can be spent quickly if the loan book turns. In the FDIC’s wording, noncurrent loans are loans 90 or more days past due or on nonaccrual. The past-due and nonaccrual rate is wider, because it also counts loans that are late but not yet in that noncurrent bucket. Many stock write-ups say nonperforming loans. Treat that label as a prompt to read the footnote, not as a guarantee that the definition matches the US one.
Loans That Count as Noncurrent
Call report instructions put a loan on nonaccrual in any one of three cases: the bank has moved it to a cash basis because the borrower’s finances have worsened, full payment of principal or interest is no longer expected, or principal or interest has been in default for 90 days or more unless the asset is both well secured and in the process of collection. A loan secured by a 1-to-4 family home, and many consumer loans, need not go on nonaccrual for lateness alone, so a clean nonaccrual line can still hide late retail loans that are reported elsewhere.
- The stock: Noncurrent loans, or the local nonperforming figure, divided by total loans.
- The flow: Net charge-offs, the losses the bank has actually taken, divided by loans.
- The direction: Both figures against the prior year, not against a single quarter you liked.
A concentration belongs in the same reading. If one category, such as commercial real estate or loans to other financial firms, is a large share of the book, a problem in that market is a problem for the share. Growth that is much faster than peer banks in that same category is the management question hiding inside the credit table.
The Reserve Against Those Loans
The allowance for credit losses is the reserve set aside for loans that may not be repaid. Coverage, in the FDIC profile, is that allowance divided by noncurrent loans, not by every loan that is a few weeks late. A quarter with a small new provision can still be a weak quarter if problem loans rose and coverage fell. Industry marks from the FDIC second quarter 2026 banking profile are a backdrop for US banks, not a pass mark for the one you are buying.
| Credit mark | All FDIC-insured banks, second quarter 2026 |
|---|---|
| Past-due and nonaccrual rate | 1.44% |
| Quarterly net charge-off rate | 0.57% |
| Allowance divided by noncurrent loans | 172.7% |
| Same coverage ratio at community banks | 145.6% |
The release describes industry asset quality as improved in that quarter, and it also says some loan portfolios were still weak. Your bank can sit on the wrong side of both statements. If its noncurrent ratio is rising and coverage is falling toward the loans, the reserve is the number to trust, not the headline that provisions were light.
Margin, Costs, and Price to Book
Once capital and credit can carry the dividend, the next question is whether the bank earns enough on that equity to deserve the price. Net interest margin is net interest income divided by average earning assets. It is the spread after the bank has paid for deposits and other funds. Fee income sits beside it. In the second quarter of 2026, higher noninterest income was a main driver of the industry’s earnings increase, so a bank that lives on spread alone is not the same investment as a bank with a large fee line.
Building a Net Interest Margin
In the second quarter of 2026, the net interest margin for all FDIC-insured banks was 3.32 percent. That single industry figure is a poor target for every bank, because business mix and size change the spread.
Community banks as a group posted 3.81 percent in that same quarter. Compare a small lender with other small lenders, and a giant with other giants. A wider margin is not free: it can mean riskier borrowers or a loan book that reprices slowly when deposit rates jump.
Return on assets for the industry was 1.37 percent in that quarter. Return on equity will look larger than return on assets because the equity slice is thin, which is why a high return on equity can be leverage rather than a wide spread. The efficiency ratio, noninterest expense divided by revenue, tells you the cost of producing that revenue. A bank that spends 60 cents to produce a dollar of revenue has an efficiency ratio of 60 percent. There is no universal pass mark; the comparison is banks of a similar size and mix.
Price to Book From a Return on Equity
Bank shares are usually priced against book value, because the assets and liabilities are mostly financial claims. Price to book is the share price divided by book value per share. Price to tangible book subtracts goodwill and other intangibles first. Goodwill from old takeovers cannot absorb loan losses the way tangible equity can, so a bank that has bought other banks often looks cheaper on plain book than on tangible book.
A WORKED PRICE TO BOOK
This illustration uses inputs you can swap for your own. Assume the bank earns a 12 percent return on equity, you require 10 percent a year for the risk, and book value grows at 3 percent. The justified price to book is (12 percent minus 3 percent) divided by (10 percent minus 3 percent), which is 0.09 divided by 0.07, or 1.29 times book. If the shares trade well above that, you are paying for a return that has to stay higher, or for growth that has to stay faster. If they trade well below it, the market is pricing a lower return, or losses that will cut the book.
| Illustration per share | Amount |
|---|---|
| Share price | $36 |
| Book value | $30 |
| Goodwill and other intangibles | $6 |
| Tangible book | $24 |
| Price to book | 1.2 times |
| Price to tangible book | 1.5 times |
The same $36 share is 1.2 times book and 1.5 times tangible book. Neither multiple is the whole decision. Set the multiple next to the return on tangible equity, and next to the capital and credit checks, before you call a price under book a bargain. A discount is often the market’s estimate that future losses, or a return below what owners require, will shrink what the book is worth.
Deposits, Rate Marks, and Off-Balance-Sheet Risk
Liquidity and market risk are the checks that do not always run through this quarter’s profit. You want deposits that stay, a loan book that is not racing ahead of those deposits, and a bond book whose market value you have actually read. Unused credit lines, guarantees, and other commitments are real exposures on the day a customer draws them, even while they sit off the balance sheet. A state backstop, where one exists, is support for the system and for depositors, not a floor under the share price.
RATE MARKS IN THE SECOND QUARTER OF 2026
- Unrealized losses: Losses on held-to-maturity and available-for-sale securities were $326.7 billion, up from $325.1 billion the prior quarter.
- Share of cost: Those losses were 5.5 percent of amortized cost, compared with 6.8 percent a year earlier.
- Long-dated assets: Loans and securities longer than three years were 33.0 percent of industry assets, the lowest share since 2014, and 43.2 percent at community banks.
The loss figure is not a prediction that every bank will sell those bonds and take the hit. It is the gap between book value and market value on the securities book, and it is larger, as a share of assets, at many smaller banks because more of their balance sheet is long-dated. If your bank’s tangible book ignores a large unrealized loss on bonds it says it will hold, the price-to-book math in the previous section is using a book the market may not believe.
Domestic deposits at insured banks had risen for an eighth straight quarter by then, which describes the industry, not your bank. On the bank you are buying, compare loan growth with deposit growth, and read how much funding comes from brokers or short-term wholesale markets. A bank that must replace a rush of departing deposits will pay up, and the margin you liked can compress in a single quarter.
Checklist Before You Buy a Bank Share
Run this on the latest report before you look at the yield. The numbers in the capital table are the US base rule. If the bank is supervised somewhere else, swap in the home regulator’s minimum and its name for bad loans, then keep the same sequence.
SEVEN CHECKS ON ONE BANK
- Capital stack: Write down CET1, tier 1, total capital, and leverage, plus any higher minimum the bank says it must meet.
- Payout gap: Subtract the regulatory CET1 minimum from reported CET1. If that gap is not larger than the conservation buffer, plus any extra buffer, the dividend and buybacks can be capped.
- Late loans: Write the noncurrent or nonperforming ratio and the net charge-off rate, and note whether each is higher than a year ago.
- Reserve: Divide the allowance by noncurrent loans. A falling coverage ratio while late loans rise is the stop sign.
- Spread and funding: Read the net interest margin, the share of revenue from fees, and whether deposits are growing with the loans.
- Price: Divide the share price by tangible book per share, and set it next to the return on equity using the worked formula above.
- Hidden lines: Note any single loan concentration, unused commitments, and unrealized losses on bonds. Only then read the dividend yield.
This fits an investor who can hold through a credit cycle and who will reopen the report every quarter. It is a poor fit if you need the cash within a year or two, or if one bank would be a large share of your savings. A high yield does not shorten the cycle. The loan book still has to season, and the bond marks move with rates whether or not you wanted the income.
Common Mistakes With Banking Shares
The mistakes below are the ones that survive after the checklist, because each one uses a number that looks like an answer.
MISTAKES THAT SKIP THE LOAN BOOK
- The yield: Sorting banks by dividend yield puts the banks whose prices have already fallen at the top of the list.
- The discount: A price under book, without the credit trend and the bond marks, is often the loss written into the quote.
- Goodwill: Book value after a string of takeovers includes intangibles that do not absorb loan losses.
- A quiet quarter: A small provision, or a release of old reserves, can flatter earnings while problem loans are rising.
- Deposit insurance: Insurance is written for depositors. It does not protect the share price if equity is used up.
- Ordinary ratios: EBITDA does not fit a bank, because interest expense is the cost of the product, and a raw price-to-earnings ratio skips capital and credit.
- The wrong peer: A global bank’s margin and a community bank’s margin are not one comparison.
Two of these together are the pattern to walk away from until the next report: a discount to tangible book, and a reserve that is not keeping up with late loans. A long history of dividends does not cancel that pair. The dividend is a decision the board can revisit as soon as the capital gap narrows.
Frequently Asked Questions
Where Do You Find Capital and Loan Figures for a Bank?
Start in the annual report and the latest quarterly earnings release, with the capital ratios and the credit-quality tables, not the dividend headline. A US bank also files a call report, and the FFIEC Uniform Bank Performance Report lines those regulatory figures up for comparison. If the only number in an investor slide is the yield, you do not have the case yet.
How Often Should You Recheck Banking Shares You Own?
Run the same checks every time the bank reports earnings, and again when it changes the dividend, announces a buyback, or flags fast growth in one loan type. Waiting for the annual report can skip a quarter in which late loans rise and the reserve does not follow. Put four figures beside the prior quarter: capital, late loans, coverage, and the margin.
Do Bank Share Checks Change Outside the United States?
The risks are the same, and the legal minimums are not. Use the capital rule and the bad-loan definition from the bank’s home regulator, then judge the bank against local peers of a similar size. Names such as NPA or NPL are local labels for problem loans, and a ratio that looks comfortable under one country’s rule can sit close to another’s penalty line.
Is One Strong Quarter Enough to Buy Banking Shares?
One strong quarter is not a buying case. You want several quarters in which capital stays clear of the payout limit, late loans are stable or falling, and the reserve still covers them. A quarter lifted by a securities gain, or by a release of old provisions, can look clean while the loan book is not.
The next earnings release is the test of a bank share you already like. If capital is still clear of the payout limit, late loans are not outrunning the reserve, and the margin is not a one-time gift from the securities book, the dividend has a base for another quarter. If one of those three slipped, the yield you were shown at purchase is a different promise than the one in the latest report.
Disclaimer: This guide is educational and informational. It is not investment advice, a recommendation to buy or sell any banking share, or a substitute for advice from a licensed adviser who knows your finances. Capital rules, bad-loan definitions, and a bank’s reported figures change, and the US thresholds here reflect the cited rules and the FDIC profile for the second quarter of 2026. Consult a qualified financial adviser, and the bank’s own filings, before you act.
-
NEWS4 months agoMicrosoft’s 96% Cyber Score Sits at 86.3% on the Board
-
BUSINESS6 months agoMusk’s $914 Billion Lead Is a Public SpaceX Bet
-
ENTERTAINMENT1 year agoSkrilla’s 6-7 Became Word of the Year by Meaning Nothing
-
SPORTS4 months agoFree Live Sports Streaming Without Cable: An Antenna and App Guide
-
GAMING2 months agoXbox Puts 15-Hour Caps on Game Pass Cloud Play
-
ENTERTAINMENT5 months agoEndgame Encore’s $86 Million Trial for Infinity Vision
-
BUSINESS2 years agoRetail Banking vs Commercial Banking: Who Each Side Serves
-
ENTERTAINMENT2 years agoAndrew Garfield’s Spider-Man Films Are No Longer Free
